rewire.it
Task

BEACON SSP: Secondary structure prediction

Secondary structure prediction. Multi-label Cls at nucleotide level, scored with F1. Dataset bpRNA-1m; train/validation/test 10,814/1,300/1,305.

17 evaluations · 17 metric rows

Overview

Secondary structure prediction. Multi-label Cls at nucleotide level, scored with F1. Dataset bpRNA-1m; train/validation/test 10,814/1,300/1,305.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Evaluation design

Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.

Benchmarks

These source-backed links do not make different protocols or scores interchangeable.

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Explore all linked results

Run instructions

No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.

Published comparisons

Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.

BEACON SSP: Secondary structure prediction

f1 (percent) · Higher values are better for this metric.

Every method in BEACON Table 3 on Secondary structure prediction, scored with F1 on bpRNA-1m with the split 10,814/1,300/1,305.

Evaluation protocol · bpRNA-1m (BEACON split)

  1. CNN · Method · Author-reported evaluation49.95(0.82)
  2. ResNet · Method · Author-reported evaluation57.26(3.14)
  3. LSTM · Method · Author-reported evaluation58.61(0.21)
  4. RNA-FM · Configuration · Author-reported evaluation68.50(0.54)
  5. RNABERT · Configuration · Author-reported evaluation57.27(0.30)
  6. RNA-MSM · Configuration · Author-reported evaluation57.98(0.47)
  7. Splice-H510 · Configuration · Author-reported evaluation64.93(0.84)
  8. Splice-MS510 · Configuration · Author-reported evaluation43.24(28.64)
  9. Splice-MS1024 · Configuration · Author-reported evaluation68.26(0.20)
  10. UTR-LM-MRL · Configuration · Author-reported evaluation59.71(0.30)
  11. UTR-LM-TE&EL · Configuration · Author-reported evaluation59.57(0.20)
  12. UTRBERT-3mer · Configuration · Author-reported evaluation60.37(0.47)
  13. UTRBERT-4mer · Configuration · Author-reported evaluation59.41(0.45)
  14. UTRBERT-5mer · Configuration · Author-reported evaluation47.92(8.75)
  15. UTRBERT-6mer · Configuration · Author-reported evaluation38.56(28.76)
  16. BEACON-B · Configuration · Author-reported evaluation64.18(0.44)
  17. BEACON-B512 · Configuration · Author-reported evaluation58.75(3.72)

Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.

BEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,column(SSP)
Values, uncertainty and evidence
f1: original source values
Tested entityPrinted valueUncertaintyEvidence
CNN · Method49.95(0.82) percenttype: standard_deviation; value: 0.82Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(CNN),column(SSP)
ResNet · Method57.26(3.14) percenttype: standard_deviation; value: 3.14Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(ResNet),column(SSP)
LSTM · Method58.61(0.21) percenttype: standard_deviation; value: 0.21Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(LSTM),column(SSP)
RNA-FM · Configuration68.50(0.54) percenttype: standard_deviation; value: 0.54Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNA-FM),column(SSP)
RNABERT · Configuration57.27(0.30) percenttype: standard_deviation; value: 0.30Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNABERT),column(SSP)
RNA-MSM · Configuration57.98(0.47) percenttype: standard_deviation; value: 0.47Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNA-MSM),column(SSP)
Splice-H510 · Configuration64.93(0.84) percenttype: standard_deviation; value: 0.84Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-H510),column(SSP)
Splice-MS510 · Configuration43.24(28.64) percenttype: standard_deviation; value: 28.64Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-MS510),column(SSP)
Splice-MS1024 · Configuration68.26(0.20) percenttype: standard_deviation; value: 0.20Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-MS1024),column(SSP)
UTR-LM-MRL · Configuration59.71(0.30) percenttype: standard_deviation; value: 0.30Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTR-LM-MRL),column(SSP)
UTR-LM-TE&EL · Configuration59.57(0.20) percenttype: standard_deviation; value: 0.20Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTR-LM-TE&EL),column(SSP)
UTRBERT-3mer · Configuration60.37(0.47) percenttype: standard_deviation; value: 0.47Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-3mer),column(SSP)
UTRBERT-4mer · Configuration59.41(0.45) percenttype: standard_deviation; value: 0.45Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-4mer),column(SSP)
UTRBERT-5mer · Configuration47.92(8.75) percenttype: standard_deviation; value: 8.75Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-5mer),column(SSP)
UTRBERT-6mer · Configuration38.56(28.76) percenttype: standard_deviation; value: 28.76Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-6mer),column(SSP)
BEACON-B · Configuration64.18(0.44) percenttype: standard_deviation; value: 0.44Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(BEACON-B),column(SSP)
BEACON-B512 · Configuration58.75(3.72) percenttype: standard_deviation; value: 3.72Author-reported evaluation · source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(BEACON-B512),column(SSP)
Scope and limitations
  • Author-reported numbers, source checked but not independently reproduced.
  • The Literature SOTA row is excluded: those numbers come from other papers under their own protocols.
  • Metrics differ between tasks, so these figures cannot be averaged into one RNA score.

Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.

Tested entities and results

Release 2026-09-17-1a18ca2c038a · 17 evaluations · 17 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.

Results grouped by the exact reported evaluation
Metric and findingCoverage and uncertaintyEvidence
BEACON-B on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

64.18(0.44)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.44

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(BEACON-B),column(SSP)

Source checking is not independent reproduction.

BEACON-B512 on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

58.75(3.72)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 3.72

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(BEACON-B512),column(SSP)

Source checking is not independent reproduction.

CNN on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

49.95(0.82)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.82

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(CNN),column(SSP)

Source checking is not independent reproduction.

LSTM on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

58.61(0.21)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.21

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(LSTM),column(SSP)

Source checking is not independent reproduction.

ResNet on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

57.26(3.14)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 3.14

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(ResNet),column(SSP)

Source checking is not independent reproduction.

RNA-FM on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

68.50(0.54)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.54

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNA-FM),column(SSP)

Source checking is not independent reproduction.

RNA-MSM on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

57.98(0.47)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.47

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNA-MSM),column(SSP)

Source checking is not independent reproduction.

RNABERT on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

57.27(0.30)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.30

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(RNABERT),column(SSP)

Source checking is not independent reproduction.

Splice-H510 on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

64.93(0.84)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.84

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-H510),column(SSP)

Source checking is not independent reproduction.

Splice-MS1024 on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

68.26(0.20)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.20

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-MS1024),column(SSP)

Source checking is not independent reproduction.

Splice-MS510 on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

43.24(28.64)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 28.64

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(Splice-MS510),column(SSP)

Source checking is not independent reproduction.

UTR-LM-MRL on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

59.71(0.30)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.30

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTR-LM-MRL),column(SSP)

Source checking is not independent reproduction.

UTR-LM-TE&EL on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

59.57(0.20)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.20

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTR-LM-TE&EL),column(SSP)

Source checking is not independent reproduction.

UTRBERT-3mer on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

60.37(0.47)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.47

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-3mer),column(SSP)

Source checking is not independent reproduction.

UTRBERT-4mer on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

59.41(0.45)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 0.45

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-4mer),column(SSP)

Source checking is not independent reproduction.

UTRBERT-5mer on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

47.92(8.75)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 8.75

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-5mer),column(SSP)

Source checking is not independent reproduction.

UTRBERT-6mer on BEACON SSP: Secondary structure prediction

BEACON harness, fixed downstream head per task. Train/validation/test 10,814/1,300/1,305; dataset bpRNA-1m.

Author-reported evaluation · Evaluation metadata: source checked

38.56(28.76)% f1

Unit: percent · Direction: higher

Uncertainty: type: standard deviation; value: 28.76

Scored: Not reported · Eligible: Not reported

source checkedBEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2) · Table3,p.8,row(UTRBERT-6mer),column(SSP)

Source checking is not independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-1a18ca2c038a
Property and statementOriginal source and locationReview and provenance
Relationship: part of

discovery-benchmark-beacon

Individual claims
BEACON: Benchmark for Comprehensive RNA Tasks and Language Models (arXiv:2406.10391v2)

Original source ↗

Table1,p.6,row(SSP)

Version: v2, 12 December 2024
Retrieved: 2026-09-18

source checked

automated source review · 2026-09-18

Audit details

Primary-source transcription with no human sign-off and no independent reproduction.

Field: links:part_of:discovery-benchmark-beacon

Claim: beacon-association-ssp

Source artifact SHA-256: 1370d75fe591bb8f994bc67f100a1a3fd557a62b9edb962b21f2fb038e9dea16

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Sources and history

Release 2026-09-17-1a18ca2c038a · Record review: source checked

1 source records and release historyDownload this release
Technical metadata and extraction receipts

Stable ID: beacon-task-ssp

areas
rna-transcriptomes
tasks
Secondary structure prediction
metric
F1
metric direction
higher
task type
Multi-label Cls
rna level
Nucleotide
dataset
bpRNA-1m
splits
10,814/1,300/1,305
source locator
Table1,p.6,row(SSP)
comparison panels
id: beacon-panel-ssp; title: BEACON SSP: Secondary structure prediction; protocol id: beacon-task-ssp; dataset id: beacon-dataset-bprna-1m; metric: f1; unit: percent; direction: higher; result ids: beacon-result-cnn-ssp-f1; beacon-result-resnet-ssp-f1; beacon-result-lstm-ssp-f1; beacon-result-rna-fm-ssp-f1; beacon-result-rnabert-ssp-f1; beacon-result-rna-msm-ssp-f1; beacon-result-splice-h510-ssp-f1; beacon-result-splice-ms510-ssp-f1; beacon-result-splice-ms1024-ssp-f1; beacon-result-utr-lm-mrl-ssp-f1; beacon-result-utr-lm-te-and-el-ssp-f1; beacon-result-utrbert-3mer-ssp-f1; beacon-result-utrbert-4mer-ssp-f1; beacon-result-utrbert-5mer-ssp-f1; beacon-result-utrbert-6mer-ssp-f1; beacon-result-beacon-b-ssp-f1; beacon-result-beacon-b512-ssp-f1; source ids: source-beacon-arxiv-2406-10391; source locator: Table3,p.8,column(SSP); context: Every method in BEACON Table 3 on Secondary structure prediction, scored with F1 on bpRNA-1m with the split 10,814/1,300/1,305.; caveats: Author-reported numbers, source checked but not independently reproduced.; The Literature SOTA row is excluded: those numbers come from other papers under their own protocols.; Metrics differ between tasks, so these figures cannot be averaged into one RNA score.; review: method: automated_source_review; date: 2026-09-18
Related records

Suggest a correction